panda-fail-run

name	panda-fail-run
description	Send a terminal FAILED callback when the run cannot continue for a technical or execution reason that is not missing external input
compatibility	opencode
metadata	{"callback":"FAILED"}

What this skill does

• Use panda_fail_run
• End the current run in failure
• Provide a concise failure summary

Use this skill when

• a validation command fails and you cannot resolve it
• the codebase or environment is broken in a way that blocks progress
• the run cannot continue for a technical reason

Do not use this skill when

• the real problem is missing external information
• the run is actually complete
• the code change is finished but automated tests are absent, unsupported, or no tests match the changed area
• you can describe the remaining risk as a validation gap in panda_complete_run

Required payload

• summary

Optional payload

• details

Quality bar

• explain the failure precisely
• include the failing command or subsystem in details when useful
• reserve this skill for real technical dead-ends, not missing or unavailable tests
• never include source code, diffs, credentials, or secret values in the failure payload